The Frontier Red Team (Anthropic) used the Claude language model in collaboration with Theorem Labs and, within a few weeks, proved security properties in more than a hundred real, open-source libraries. The work indicates that AI can increase the scalability of formal verification, which could improve software security.
AI-generated text
Large-scale formal software verification by Frontier Red Team using Claude
The Frontier Red Team (Anthropic) used the Claude language model in collaboration with Theorem Labs and, within a few weeks, proved security properties in more than a hundred real, open-source libraries.
